Multiple sites for preregistering studies exist. The Center for Open Science (https://www.cos.io/initiatives/prereg) and the Open Science Foundation (https://help.osf.io/article/330-welcome-to-registrations) have sites open to a wide range of disciplines, including political science, and the American Economic Association offers a site (https://www.socialscienceregistry.org) for preregistering randomized controlled trials. Journals in political science increasingly require preregistration for many types of studies. If you have any plausible expectation of publishing your work in a professional (as opposed to student) journal, you should explore whether preregistration makes sense for you. It does constrain your ability to explore your data, at least without further data collection beyond the set that you registered, but it increasingly constrains your ability to publish. If preregistration might be for you, the Open Science Foundation offers a template at https://doi.org/10.31222/osf.io/epgjd that will help you draft a preregistration document.
Registered reports are a way to ensure publishability of potential null results (non-findings). In this form of publication, the authors submit an extended preregistration document containing more literature review and theory than they otherwise might, along with hypotheses, data collection methods, and intended forms of analysis. Critically, this stage happens before data collection and analysis. The preregistration document then undergoes results-blind peer review — reviewers see only the preregistration document, not the actual results (which the authors have not yet obtained either). The goal is to determine whether the study itself merits publication, regardless of the actual results. Non-findings matter, and this is a way to help avoid the inherent publication bias of reviewers (and authors, to be fair) preferring to publish only affirmative results. If reviews are positive, a journal may choose to issue a form of conditional acceptance to the authors: If the research is completed as planned, and paper passes a traditional peer review, the journal agrees to publish the paper no matter the findings.
Good resources on preregistration in political science include:
- The APSA’s Experimental Politics section’s guide: https://connect.apsanet.org/s42/2021/04/28/what-is-a-registered-report-and-why-should-you-be-writing-one-for-jeps/
- A more detailed guide on the process from the Journal of Experimental Political Science: https://www.cambridge.org/core/journals/journal-of-experimental-political-science/information/faqs-for-registered-reports
- The Journal of Politics‘ guide: https://www.journals.uchicago.edu/journals/jop/registered-report-guidelines